Proposal: Budget OB, inspired by Paul's Aethers

[gtforme00]

Paul's Aether design has really got me thinking about this configuration today. It seems that the open midrange approach with a monopole bass bin is the ideal compromise for a budget dipole experience.

Since the original design was built around available surplus drivers, I have an idea for another design using the NHT surplus drivers here.

Using the 1" aluminum dome.


The 5.25" Peerless midrange.


And the 8" Peerless woofer.


The dome is similar to a Seas, most likely being selected as a lower cost alternative to a Seas unit for the lower priced NHT models. It is crossed over to between 2.4 and 2.7kHz in the Super Audio series.

The mid-woofers have distortion reducing shorting rings and large (4.5mm) x-max. They were crossed over at 2.5kHz in the vt-2.4. They seem like they would be easy enough to match up to the 1" dome tweeters.

The woofers were used in a vented enclosure of less than 1.5 cu ft size in the NHT 2.5i. NHT claimed -3db at 29Hz which is similar to Paul's design. The woofers also have similar excursion to the Dayton DVC 8s.

Looking at the crossover points for the original speakers in which the drivers were used, it seems that the ideal crossover point would be around 2.5k 12db/octave.

One could get all of those drivers, including shipping for just over $100. I bet with careful selection of crossover components the total cost including construction materials could come under $200.

Just ruminating aloud,
David

[Andy G]

"It seems that the open midrange approach with a monopole bass bin is the ideal compromise for a budget dipole experience."

That is what was intended. As I've said before, I'm really glad Paul took up the idea and is spreading it around.

As for your combination
Looks ok to me.

Do use the MTM up top (ie, not an MT), and do keep the mid-tweeter x-o point around the 2k-2.4k region, lowish order, and everything should be fine. Maybe even keep the low order on the woofer as well if possible.

[ttan98]

Originally posted by Andy Graddon Un
"It seems that the open midrange approach with a monopole bass bin is the ideal compromise for a budget dipole experience."

That is what was intended. As I've said before, I'm really glad Paul took up the idea and is spreading it around.

As for your combination
Looks ok to me.

Do use the MTM up top (ie, not an MT), and do keep the mid-tweeter x-o point around the 2k-2.4k region, lowish order, and everything should be fine. Maybe even keep the low order on the woofer as well if possible.

I have experimented INDEPENDENTLY with open baffle MTM with about 2khz x-over with tweeter. I concur with Andy's findings. Furthermore my MTM is very efficient, 95-96db but with only 4 ohm impedance. I am using active x-over,it should not be a problem. I believe if you can get the more efficiency out off MTM the better, it is easier to drive.

Cheers.

[Undefinition]

Originally posted by gtforme00

The woofers were used in a vented enclosure of less than 1.5 cu ft size in the NHT 2.5i. NHT claimed -3db at 29Hz which is similar to Paul's design. The woofers also have similar excursion to the Dayton DVC 8s.

Looking at the crossover points for the original speakers in which the drivers were used, it seems that the ideal crossover point would be around 2.5k 12db/octave.

One could get all of those drivers, including shipping for just over $100. I bet with careful selection of crossover components the total cost including construction materials could come under $200.

[/B]

Those look like great drivers. Killer price! The only thing I might balk at would be the tweeter. But I've never heard it, so what do I know? That site is weird; they tell you every technical measurement, but I don't see any FR graph.

But yeah, looks good to me, though I agree that OB mids are better when there's more surface area, even if they're run in series.

-Paul

[gtforme00]

Thanks for the votes of confidence! It was the original intent to stay with the MTM configuration, I completely agree with all of the above comments as to why. I also think that in addition to the above, the MTM configuration allows the advantage of more directivity in the vertical plane. Combined with the horizontal directivity of a dipole(ish) midrange, it seems like this configuration would have excellent power response.

Concerning the tweeter, It has been well reviewed in the NHT Super Audio series crossed over at anywhere from 2.4-2.7 kHz LR2. Jack Hidley from NHT has hinted that is is very similar to the Seas used in the VT3.

The tweeters on the sale page have 1" voice coils, ferrofluid, neo motors, fabric surround and tinsel lead out wires. When used with the heatsink, we have used virtually identical tweeters in speakers such as the VT3, which is a four way tower speaker that can play very loudly.

I have a hunch (in no way am I implying that Jack Hidley told me this) that is is more or less a clone of a Seas design (25TAFN looks close) that was second sourced to Asia for cost savings. As inexpensive as they are, one could cheaply play with dipole tweeters as well!

I'll be ordering some speakers from Jack after collecting my interest free loan back from Uncle Sam. I'm already ordering drivers for a NHT M3.3 clone, so might as well go ahead and grab enough extras to play with this design.

Regarding the lack of frequency response graphs on the webpage; I don't think that loudspeaker engineers place much value in typical frequency response graphs. With all of the other parameters to take in consideration, the raw quasi-infinite baffle response is probably near the bottom of the list of things to look at in driver selection. I think that looking at how they used the original drivers, and the frequencies at which they are crossed over will tell a lot about their suitability for a design.

Check out this picture of the NHT VT-2.4

That's the exact same midrange drivers for sale there (tweeter is similar as per above), and they are already in an MTM configuration! This is encouraging, and if we can get a copy of the VT-2.4 crossover schematic, we could be several steps ahead. The baffle width will be different, requiring some tweaking, but it is a start.

I'm very excited about this possibility!

Regards,
David

[allthumbs]

I recently found that sale page and I also think this might be a fun, inexpensive , killer project. What about the seas midrange for your m-t-m? Unfortunately I don't have any Idea how to go about designing the x-over. Does one need to run fr curves in order to predict x-over results? Every time I read a tutorial on using a computer for x-o design I get a headache.

[Andy G]

One of the ideas behind the Aether design is to use mids that have a nice smooth response, no break-up peaks etc , so that even the x-o can be pretty minimal, and preferably of low order type.
Some of the Seas mids suit this admirably, but some don't, it depends on which mid you are referring to.

[Moondog55]

I have always disliked the idea of having to pad down the most important part of the audio spectrum ( the midrange 300-300 +/- )
So I still haven't decided how to build these, the Vifa's are only 87/88dB and I hate to lose any efficiency, I may go with Rabbitz method and add a .5 driver firing to the rear or even dual .5's firing front and back in a cascaded topology.

[Jack Hidley]

To do any decent speaker design, you must measure the frequency response of the drivers in your actual cabinets. Without this data, crossover modeling is virtually worthless. Fortunately, the equipment to do this is so cheap that there is no excuse. Your system will sound so much better and you will be amazed at how much you learn in the process.

I didn't post any frequency response measurements on the website for a couple of reasons. If I take all of the measurements in our anechoic chamber, people will expect the low end and low midrange to look like that when they put the driver in their cabinet with 6dB of diffraction loss. Some will also think that the driver has a 120Hz peak in it from the chambers first mode.

If I measure them in a 4pi environment in a cabinet, then half the people will think the driver has a rising response curve. I've been down this road many times. It's better to not post anything. To do any really meaningful design, the drivers need to be measured in the actual cabinets anyway.

[Andy G]

Originally posted by Jack Hidley
To do any decent speaker design, you must measure the frequency response of the drivers in your actual cabinets. Without this data, crossover modeling is virtually worthless. Fortunately, the equipment to do this is so cheap that there is no excuse. Your system will sound so much better and you will be amazed at how much you learn in the process.

I didn't post any frequency response measurements on the website for a couple of reasons. If I take all of the measurements in our anechoic chamber, people will expect the low end and low midrange to look like that when they put the driver in their cabinet with 6dB of diffraction loss. Some will also think that the driver has a 120Hz peak in it from the chambers first mode.

If I measure them in a 4pi environment in a cabinet, then half the people will think the driver has a rising response curve. I've been down this road many times. It's better to not post anything. To do any really meaningful design, the drivers need to be measured in the actual cabinets anyway.

I understand your point of view on this, but for myself, I generally like to use low order x-os with minimal correction to the drivers, so I like to see a frequency response for drivers I might consider purchasing.

Most people understand that the lower end of woofs and mids cannot really be measured well, and varies with cabinet, baffle etc. but I want to know if mid/woofers have a niceish roll-off at its top end or if they have a big break-up peak or all over the place response BEFORE I will even consider purchasing. Same with tweeters, I want to know if they have a generally flat response or if they have a rising response etc so I have some idea what I might be in for with the x-o.

If there are no frequency curves posted on a web site , I generally just bypass those drivers unless someone else has posted measurements somewhere.

We all have our own ways of going about things I guess.